ResearchTasksMulti-turn agent decision makingMulti-turn agent decision makingFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedToolSandbox (test)H-EPM52.2Success Rate7Feb 26, 2026tau2-Bench (test)H-EPM22.3Success Rate7Feb 26, 2026tau-Bench (test)H-EPM55.8Success Rate7Feb 26, 2026